Some Aspects of a Framework for Energy Data

Little doubt exists that technologies for precisely, and automatically measuring energy use are timely. Pressure to reduce greenhouse gas emissions, the potential for a generation shortfall, and rising energy prices, modern building complexities, mean that estimating building energy use patterns, profiling of building energy use, and energy failure mode identification can help to maintain energy efficiency. However, further uses exist beyond operational management; These include urgently required meta-analysis of building stock by sector, up to national stock levels, to inform policymakers, since a dearth of national stock data exists at government level in many countries. Many existing systems use radio telemetry, often producing unclean data. Analysis of energy data for large datasets becomes expensive due to incompatible formats, hampering use of old data on new systems, and data from different systems should we acquire new data, or as we acquire physical buildings. We argue that a standard should include basic specifications for fundamentals such as date formats, but a secondary scalable layer will allow future-proofing of datasets for longitudinal study, and open the door to advanced analysis techniques such as complex event processing. Disaggregation to plant level, as well as building related activities, such as manufacturing activity, also becomes possible with a scalable data structure. This paper, proposes a framework for an energy data standard from a data analysis perspective built around four areas: Temporal, accuracy & precision, operational and energy documentation.